Identifying myself as an expat has largely shaped my personality, perspective, and aspirations. Since 2015, I have gained extensive experience in research, development and deployment of AI solutions, and leadership, fostering collaborations with diverse academic and industrial partners.

This multifaceted experience helped me form a structured approach to solving problems at work, with three fundamental steps: understanding and visualizing problems quickly, developing and implementing innovative solutions, and evaluating and optimizing the outcomes. This approach has been pivotal in my professional success and personal development. I continuously expand my knowledge and stay up to date with the latest advancements in artificial intelligence and digital health.

My mission is to address challenges in these fields, from multimodal data collection to extracting actionable insights and delivering personalized interventions, by leveraging AI, health data processing, and wearable Internet-of-Things technologies. My expertise spans designing data analysis methods – such as signal processing, machine learning, and large language models – and developing and deploying AI agents in mobile health and conversational health applications.

My job responsibilities have included the following:

  • Research: I have built theoretical knowledge across a wide range of topics, such as artificial intelligence, the Internet of Things, health data analytics, large language models, wearable electronics, edge computing, and digital health. My current focus is on LLM-based agents, agent evaluation, and safety guardrails for production health applications. My broader work has covered data analysis methods for biomedical signals, e.g., photoplethysmogram (PPG) and electrocardiogram (ECG); personalized machine-learning-based health data analytics (e.g., CNN-, LSTM-, and GAN-based); the application of LLMs in healthcare; anomaly event detection methods for longitudinal and multivariate data; and signal quality assessment, to mention a few. I am the co-creator of openCHA, an open-source LLM-powered framework that enables Conversational Health Agents (CHAs) to deliver effective healthcare services, including patient support, health coaching, and promoting patient self-awareness. I also led the development of e2epyppg, an end-to-end PPG signal processing pipeline. I have collaborated with multidisciplinary teams at Thrive AI Health, the University of California Irvine, Stanford University, iHealth Labs, University of Turku, Vienna University of Technology, and Silo AI, among others. I am the author of more than 80 peer-reviewed publications in both medical and technological venues.
  • Development and Deployment: I have built my coding skills (mostly in Python since 2015) by taking an active role in different projects, including health data collection (e.g., PPG and ECG), wearable- and cloud-based health data analysis, and LLM-based agents. My recent work has focused on AI agents in production, where I owned features end to end, including the backend, frontend, CI/CD, and monitoring. In addition, I recently developed and deployed Care Companion, a proof-of-concept AI chat companion that helps family caregivers and patients with care challenges. I gained experience working with various Python libraries and tools, including TensorFlow, scikit-learn, SciPy, pandas, LangChain, Hugging Face, and Ollama.
  • Leadership: I have co-led cross-disciplinary research initiatives with faculty across Computer Science, Nursing, and Medicine, and built partnerships with external academic and industrial organizations. I have also advised master’s and Ph.D. students working on AI solutions for healthcare applications and taught two courses on biosignal acquisition and analysis for three years.

Areas of Expertise & Interests

  • LLM Agents
  • LLM and Agent Evaluation
  • AI Agent Development and Deployment
  • Digital Health
  • Wearable-based Health Monitoring and Coaching
  • Machine Learning
  • Biomedical Signal Processing